What does “40 percent lower” mean in people?
A relative change says how much a risk moved, not how big the risk was. This converter takes the risk in the comparison group and the relative change given for it, and shows the difference per 1,000 people and the number needed to treat or harm.
The tool
How it works
The method and its sources
Absolute difference = comparison-group risk × relative change. Number needed to treat (or harm) = 1 ÷ the absolute difference, rounded up to a whole person. Both hold for the length of the study the figures came from.
An odds ratio is converted exactly: with the comparison-group risk p, the treated-group risk is OR × p ÷ (1 − p + OR × p), and the result shows the risk ratio that gives. A hazard ratio How often an event happened in the treated group compared with the control group over the same period. A ratio of 0.80 means 20 percent fewer events. is treated as a risk ratio, which is close when the event is uncommon over the study period and drifts when it is common; the result says so.
This is arithmetic on the two numbers entered. It does not check them against any study, and a converter cannot tell whether the study behind them was a good one. Each EBL record shows its own absolute effects where the trial papers give them.
What happens to your data
Nothing you type leaves this device. The numbers are worked out in your browser, nothing is sent to EBL or anyone else, and nothing is saved: closing the page clears them.